Archive | IIoT

33

4:23 pm
April 25, 2017
Print Friendly

Vibration Machine Learning from the Industrial Internet Consortium

machine learning architecture for CNC machines

Figure 1: The elements and the connectivity being utilized to develop and provide updates to the production system.

Some industry analysts aren’t happy with overused buzzwords like “machine learning” or even “deep machine learning” taking the place of “IIoT” in the hype category. I agree these new buzzwords are ubiquitous in many media corners and deep machine learning is mostly found in R&D.

However, a white paper or deep dive is a great way to see what is possible for predictive analytics in the field or factory. A new white paper from the Industrial Internet Consortium, titled, “Making Factories Smarter Through Machine Learning,” offers a great read on how machine learning can allow for better edge analytics, reduce data streams and promote better data fidelity.

The white paper examines the ability of CNC machines to reduce data streams via machine learning with the use of the Plethora IIoT platform and system-on-chip engineering (SoC). The SoC technology allows for customized software to create application-specific requirements, such as data filtering unneeded data from machines.

A passage from the White Paper below:

The other capability provided by the software is the ability to read complex sensors and perform pre-processing in terms of data reduction: For example, vibration is sampled at least two times the vibration frequency. In this case, a fast Fourier transform is performed and only the frequency of interest is stored. This is an area where there is high opportunity for more efficient processing – effectively using machine learning for pre-processing and feature selection.

Therefore, it (SoC) can sample each variable with smart criterions: For example, temperature may not be measured with the same frequency of vibration

The white paper provides a real roadmap solution on how to move from preventive, SoC machine learning and simple industrial networking solutions to make this happen. The link to the white paper can be found here.

1601Iot_logoFor more IIoT coverage in maintenance and operations, click here! 

56

7:50 pm
April 14, 2017
Print Friendly

Engine OEM Identifies New Business Service

160720catlogoDisruption is an overused word in technology, but Joe Barkai’s tagline to his book about IIoT says it all: How the Industrial Internet of Things is Changing Every Business. For Mak, a supplier of engines to the maritime industry, that means changing their business model to focus and recognize that servicing their large engines remotely isn’t some wild science fiction fantasy. It’s a reality for OEMs as end users move toward IIoT strategies.

The maritime engine supplier is partnering with Caterpillar Marine Asset Intelligence (www.cat.com) and will create a condition monitoring approach for the first project. This project includes an M46 DF dual-fuel engine and will provide real-time monitoring on the ship.

“This effort enables operations and maintenance leaders to make better decisions using data and analytics, helping to drive reduced cost, downtime and risk,” says Ken Krooner, Technology & Operations Manager for Caterpillar Marine Asset Intelligence.

According to Caterpillar Marine, “the onboard analytics and user interface provide the onboard crew with real-time information, such as the condition of their equipment and what they should do about any potential issues.”

More importantly, the analytics software allows for multi-level reporting.

“At the highest level, there are high-level dashboards and reports which can provide a variety of graphs and data visualizations, including vessel performance curves, efficiency comparisons, custom metrics, geophysical location, says Leslie Bell-Friedel, global business mgr. at Caterpillar Marine Asset Intelligence in an interview for a company publication. At a detailed engineering level, there are simple red-yellow-green indicators for each piece of equipment that summarize the current and projected condition, as well as the ability to drill deep to understand the health and performance of a piece of equipment.”

Also, qualified data can be seen ashore, where additional automated analytics are used to analyze the data — both from an individual vessel as well as from a fleet perspective — and where experts are on hand to review the analytic output and apply their experience to it. Access to the analytics can be done via any web-based device, either onshore or remotely. At this point, there’s no app available.

Click here to read the Bell-Friedel’s interview >>

1601Iot_logoFor more IIoT coverage in maintenance and operations, click here! 

48

6:36 pm
April 13, 2017
Print Friendly

IIoT Trends: Pilot Projects And Silicon Valley

By Grant Gerke, Contributing Editor

The Internet of Things is changing the maintenance and reliability world. Keep up to date with our ongoing coverage of this exciting use of data and technology at maintenancetechnology.com/iot.

The Internet of Things is changing the maintenance and reliability world. Keep up to date with our ongoing coverage of this exciting use of data and technology at maintenancetechnology.com/iot.

The Internet of Things (IoT) presents an interesting picture with regard to the consumer and industrial segments. The former could be described as limping while industrial applications are moving towards market acceptance. 

Industrial manufacturing, due to a focus on equipment and networking standards for the past 20 years, is seeing the fruits of its labors through increasing use of predictive analytics. One trend emerging in 2017 is the implementation of smaller IIoT (Industrial Internet of Things) pilot projects that can provide quick results. HIROTEC, a Tier-One supplier to global automotive companies, is an example.

Justin Hester, senior researcher in the IoT Lab for HIROTEC Corp., Hiroshima (hirotec.co.jp, hirotecamerica.com), described his company’s pilot IIoT implementation in a Mar. 2017 National Association Manufacturing (NAM) webinar. “Another way to look at it is I want to crawl, walk, and then run,” he said. “I want to start visualizing data and don’t even want to do augmented reality or predictive analytics yet. That’s where I need to go, but the company can’t make that jump from no IoT solution to a full, augmented reality and predictive analytics tomorrow.”

For the SCRUM project, the automotive supplier and machine-tool maker started with eight CNC-based machines at its Detroit facility in 2015.

The project involved the use of Kepware KEPServerEX from PTC (Needham, MA, ptc.com) and OPC UA network solution to move variable CNC-machine control data to the ThingWorx (Exton, PA, thingworx.com) IoT platform to produce real-time data analytics for executives, plant managers, and technicians. Hester noted that the control platforms and data types varied greatly at this business unit, which was one of the reasons for the pilot project.

“Project flexibility is needed, especially when we’re talking about IIoT technology,” stated Hester. “If you have this long lead time with a pilot-project implementation, you learn what matters with the organization. If you have a long, two-year project, you’re stuck.”

HIROTEC’s SCRUM pilot project began with eight CNC-based machines at the company’s Detroit facility.

HIROTEC’s SCRUM pilot project began with eight CNC-based machines at the company’s Detroit facility.

HIROTEC’s application is quite common in industrial manufacturing, and one of the drivers for success is the OPC UA networking platform. This technology allows manufacturers to hold on to legacy equipment and use data from those machines, something that keeps CapEx costs down.

As more interoperability data solutions mature, the space is seeing more entrants, including, recently, Silicon Valley-based called Element Analytics Inc., San Francisco (elementanalytics.com). This company is targeting predictive analytics for the continuous-process industries. In Jan. 2017, it formally unveiled its Element Platform. The platform takes unstructured, operational sensor data from production “silos” to a cloud-based system, where asset models help predict equipment downtime.

Utilities (specifically wastewater operations) represent yet another segment that seems ripe for IIoT applications.

According to Jim Gillespie, co-founder of Gray Matter Systems, Pittsburgh (graymattersystems.com), a new Software-as-a-Service (SaaS) tool called ClariFind will alert utilities on sludge overflow failures and also predict thickening failures related to effluent not settling correctly.

How do such operations pay for new sensing technology, equipment, and better communication infrastructure?

In a recent blog post on TechCrunch (techcrunch.com), Gillespie pointed to “utilities selling solutions to other wastewater operations as the power industry has done.” As an example, he described how the Washington D.C. Water and Sewer Authority recently commercialized its intellectual property and provided a new revenue channel in the process.

Such reports are snapshots of industrial trends in IIoT and predictive analytics. More success stories are coming. Maybe consumer players will take note. MT

ggerke@acceleratedcontent.com

101

11:04 pm
April 6, 2017
Print Friendly

Big Data | Merge Control and Maintenance Data for Better Efficiencies

The IIoT framework shows how industrial data moves to the cloud for this Agile pilot project.

The IIoT framework shows how industrial data moves to the cloud for this Agile pilot project.

There’s an interesting Q&A post on Big Data, change culture, and control and maintenance data via a recent blog post from ARC’s IIoT Viewpoints site. The discussion includes Vish Avasarala, Co-founder of Saint Software Consultants, and Kenneth Smith, General Manager, Energy at Hortonworks.

There’s an interesting point about industrial manufacturing’s lack of flexibility, in general, and the challenges to change a work culture. I agree with this sentiment, in general, but the Agile project approach may help ignite cloud initiatives with some conservative manufacturers.

In the April Maintenance Technology print issue — and soon on maintenanctechnology.com — I write about HIROTEC’s condition monitoring pilot project that would fall under the Agile category. HIROTEC, a Tier-One supplier of exhaust systems for automotive OEMs, recently produced a six-week agile pilot project at their Michigan facility.

The goal was to show results quickly to management, albeit a small sample — eight legacy and new CNC machines.

Justin Hester, from HIROTEC, discussed the condition monitoring/cloud analytics pilot project in a recent National Association Manufacturing webinar:

HIROTEC argues, let’s do a small project. Let’s do a project that’s only six weeks long. It gets people excited. It’s not something that’s going to drone on for the next two years, where they have to devote the next two years of their life to as well as these other requests that will come in from the organization. They see light at the end of the tunnel already.

Hester added, “the Detroit pilot application was a success and HIROTEC is moving forward with IIoT pilot projects in Japan that deal with quality manufacturing.”

Click here to read the ARC interview >>

 

1601Iot_logoFor more IIoT coverage in maintenance and operations, click here! 

84

3:03 pm
April 4, 2017
Print Friendly

IIoT Offering from ABB and the B&R Acquisition

Modernization

Last month I discussed how IIoT devices and strategies are taking shape in the water and wastewater industry with a recent survey predicting a $20 billion investment for meters, data management, and analytics in the next eight years. Smart water is getting a lot of attention and some analysts think new business outcomes — see Joe Barkai podcast — are emerging, such as water municipalities  commercializing IP technology.

ABB recently announced a new, digital asset management initiative called Ability and it will aim for the utility and continuous process space, among others. The company emphasized their asset management platform and the analytical power of Microsoft’s Azure platform. The recent press release says the company is targeting many different industries, including oil and gas, water and wastewater, commercial building and even electric vehicle charging networks.

>> More || Smart Water Infrastructure Continues to Grow, but Real Challenges Persist

ABB’s press release on Ability:

ABB Ability helps customers in utilities, industry, transport and infrastructure develop new processes and advance existing ones by providing insights and optimizing planning and controls for real-time operations. The results can then be fed into control systems to improve key metrics such as factory uptime, speed and yield.

(** As I write this post, ABB just announced that it acquired B&R Automation, and will use these new assets to build on their Ability IIoT-platform and pursue factory manufacturing opportunities.)

For more information on the IIoT platform from ABB, click here.

1601Iot_logoFor more IIoT coverage in maintenance and operations, click here! 

124

8:52 pm
March 16, 2017
Print Friendly

Intelligent Water Making Strides towards Predictive Analytics

EXCEL XR metering pumps are designed for the specific chemical pumping requirements of municipal and industrial water treatment.

Last week, I ran across a Smart Water spending forecast from Bluefield Research and this week there’s an interesting post from Jim Gillespie, co-founder of Gray Matter Systems, a system integrator for cloud solutions and predictive analytics. All signs point to an increased spend in this sector for pump and motor sensors, but where will this investment come from?

According to Gillespie and his post on TechCruch, utilities may be able to sell “solutions” to other wastewater operations like the power industry has done. Gillespie cited how the District of Columbia Water and Sewer Authority has commercialized their intellectual property, giving them a new revenue channel. The water district is commercializing their water ammonia versus nitrate algorithm and selling it other treatment plants, according to Gillespie.

>> More || Smart Water Infrastructure Continues to Grow, but Real Challenges Persist

As I noted last week, new investment dollars are hard to come by but there’s are a lot of new use cases in the wastewater space, see below:

Another IIoT development, a new SaaS application that’s set to launch later this month, will calculate wastewater clarifier tank performance — providing quick analysis on a critical step in the wastewater process. The tool, called ClariFind, alerts utilities as they’re getting close to a failure before they experience it. ClariFind will predict when sludge will overflow and be released. This kind of problem causes EPA issues and fines that can run in the millions of dollars. It will also be able to predict a thickening failure, which is when the effluent doesn’t settle correctly and creates a costly sludge blanket in the tank. ClariFind is just one part of a water operations suite of productivity enhancers — solutions as a service.

Read the Full Post on TechCrunch >>


1601Iot_logoFor more IIoT coverage in maintenance and operations, click here! 

113

12:45 pm
March 8, 2017
Print Friendly

Smart Water Infrastructure Continues to Grow, but Real Challenges Persist

smart water markets

The US (39 projects) and the UK (21 projects) were the most active smart water markets during the last half of 2016. Source: Bluefield Research

By Grant Gerke, Contributing Writer, IIoT

A new report from Bluefield Research suggests that a massive smart infrastructure buildout is coming to the water and wastewater industry in the next eight years, with more than $20 billion to be spent in metering, data management, and analytics.

As devices, sensors and cloud solutions become cheaper over the next ten years, there will be a solid investment in this space but the research rings a little hollow to me. The U.S. industry, in particular, is aging and resources are limited but the big challenge may be in the area of system integrators. In a feature article from a couple years ago, I interviewed Roger Knutson, public works director at the biggest water and wastewater department in Minnesota. For Knutson, the real challenge was in overseeing software and plant monitoring upgrades to multiple plants with his own internal staff. System integrators weren’t in the budget.

“So, the real challenge is to maintain the different technologies during that timeframe,” says Knutson. We’re talking about the new and old versions of software running side-by-side at different plants or just at different plants.”

Even the Bluefield research report says that “a significant hurdle will be integrating legacy systems with new software platforms.” However, the challenge may be workflow processes, the less glamorous side of the asset management and IIoT narrative.

Other highlights from the research include:

• Halving non-revenue water– leaks and billing errors– and reducing energy consumption from 20% to 40%.

• The smart water sector is expected to scale to $12 billion in the US and $11 billion in Europe by 2025. Other hotspots for smart water activity include Australia, Singapore and Israel, where water stress and established utility network operators are more receptive to advanced technology adoption.

• European utilities are at the forefront of smart water in terms of operational solutions, while the US leads in terms of metering.

1601Iot_logoFor more IIoT coverage in maintenance and operations, click here! 

262

6:49 pm
February 28, 2017
Print Friendly

Process Operators and Tools May Bridge the Gap to Predictive Maintenance

170228pmartin

Peter Reynolds, contributing analyst for ARC Advisory Group.

Jim Wentzel, dir of Global Reliability at General Mills has been on the conference circuit recently and has been discussing “contextuality” when it comes to manufacturing data in the food industry. In his discussions, Wentzel discusses General Mills “data journey” as a company — their own plants and contract manufacturing plants outside the enterprise — and is pushing for data transparency throughout the entire enterprise eco-system. That means various types of plant and enterprise data, such as plant floor , instrument, machine vibration, supply chain and even other plants mixed together to make efficient decisions.

That means a lot of business units — and external companies per Wentzel— coming together and possible changes in workforce responsibilities. One scenario would be to have process operators provide key insights on equipment health due to a better working knowledge and lifecycle history of a particular asset.

>> View More | Silicon Valley Company Joins the Predictive Maintenance Party

Peter Reynolds, contributing analyst for ARC Advisory Group discusses this scenario with his most recent post, “Predictive Maintenance or Predictive Operations?” Reynolds describes how operations can lean on better tools, processes and how condition-based monitoring goes only so far:

Both Prognostics and Condition-based monitoring are still reactive approaches and have been used widely for decades. Still, many companies struggle with making significant improvements in predicting failures and extending the life of critical assets.

He goes on to write:

Therefore one might come to the conclusion that any predictive maintenance or asset reliability strategy might begin with an overarching operations strategy and weigh heavily on the skills of the process engineer. The process engineer (and not the maintenance and reliability engineer), has the ability to interpret the process data across the spectrum of the process and any assets.

The rub is that operations, maintenance and even IT need to view enterprise via data in one IIoT platform, such as ThingWorx, Element Analytics, or many other offerings that can provide varying analytics to different groups.

>> To read the full post, click here

1601Iot_logoFor more IIoT coverage in maintenance and operations, click here! 

Navigation